Voxtral 3 (TTS and transcription models)

Mistral released the Voxtral 3 audio family on October 20, 2026: four speech-to-text and text-to-speech models, all in public preview.

Lab
Mistral
Release date
Status
Public preview
API model ID
voxtral-mini-tts-3-0, voxtral-mini-transcribe-3-0, voxtral-mini-transcribe-realtime-3-0, voxtral-small-transcribe-3-0

The Mistral changelog entry of October 20, 2026, lists Voxtral TTS 3 (voxtral-mini-tts-3-0), text-to-speech with zero-shot voice cloning, streaming and contextual delivery; Voxtral Mini Transcribe 3 (voxtral-mini-transcribe-3-0), offline transcription and diarization with timestamps; Voxtral Mini Transcribe Realtime 3 (voxtral-mini-transcribe-realtime-3-0), real-time transcription; and Voxtral Small Transcribe 3 (voxtral-small-transcribe-3-0), offline transcription and diarization with timestamps. Mistral says all four are in public preview. The same day, Mistral released Voice Agents, a managed real-time voice pipeline.

Modalities
Speech to text (offline and real-time, diarization) and text to speech
Availability
Mistral API (public preview).
Official source
docs.mistral.ai
Last checked

Questions

When was Voxtral 3 released?
October 20, 2026.
Which models are in Voxtral 3?
Voxtral TTS 3, Voxtral Mini Transcribe 3, Voxtral Mini Transcribe Realtime 3 and Voxtral Small Transcribe 3.
Is Voxtral 3 generally available?
No; Mistral lists all four models as public preview.

Source

Release dates, model IDs and limits come from each lab’s own announcement, changelog, documentation or deprecations page, linked on every model with the date it was last checked. How we pick and check sources: editorial standards.

The Model Press

What are you looking for?

Search by headline, topic or keyword.